Soil Stiffness 4

[damo74]

I am designing a raft foundation using finite element analysis software. I need to insert a value for the stiffness of the bedding material (ie the soil). Could anyone give me some advice on how I calculate this value? I have heard that it is the allowable bearing pressure divided by teh amount the raft will settle, (ie 150kN/m2 / 0.01m = 15000 kN/m3) Does this make sense to anyone?

 

[FalsePrecision]

What you mean is soil stiffness modulus - for a given pressure, the soil wil displace "x" amount. This is supplied by the Geotech engr, or you can estimate it based on some Codes.

 

[miecz]

15000 kN/m3 is a reasonable value for medium sand or soft clayey soil. Typical values for modulus of subgrade reaction can be found in some soil mechanics textbooks. Bowles shows values from 10000 to 25000 for loose sand. I've used 30000 kN/m3 for a finite element analysis and found that the results (bearing pressures, bending moments) are not terribly sensitive to modulus of subgrade reaction. If you try 5000 and 50000, your results probably won't change much.

 

[dbuzz]

Bowles give the following table of approximate ranges for the modulus of subgrade reaction:

Sandy soils
Loose sand 4,800 - 16,000 kN/m^3
Medium dense sand 9,600 - 80,000
Dense sand 64,000 - 128,000
Clayey medium dense sand 32,000 - 80,000
Silty medium dense sand 24,000 - 48,000

Clayey soils
q < 200 kPa 12,000 - 24,000
200 < q < 800 kPa 24,000 - 48,000
q > 800 kPa > 48,000

Bowles, J.E., Foundation Analysis and Design, McGraw Hill Book Co, 1997

 

[JAE]

damo74 - one thing to remember is that these ranges given in textbooks aren't always what you will ultimately get at the jobsite. A good practice for designers using FEA with soil springs is to actually use the ranges given in two separate FEA analyses.

When using a high end stiffness for the soil, the soil will be relatively stiff compared with the concrete. This will tend to give you small deflections in the soil, and higher concentrations of pressure on the soil.

When using a low end stiffness for the soil, the soil will be relatively less stiff than the concrete. The concrete will then show higher stresses, moments, etc. and the soil pressure will be more uniform across the section.

This then gives you a sort of lower bound and upper bound solution - and by using both to create your design, assures you of a more confident design.

 

[damo74]

I'm designing a raft foundation for Semi-D domestic dwellings. The contractor wants to use this form of construction, only if we can design the raft for a maximum thickness of 200mm. He does not want to use ground beams. It is only economical if we use mesh top and bottom and u-bars around the edges. The ground is good with an expected bearing capacity of 150kN/m2.I have analysed the raft structure using FEA and it will work theoretically. The maximum deflection is about 5mm at the corners. I'm worried about differential settlement and excessive cracking of the blockwork. Has anybody designed a raft foundation this thin without ground beams? Has anyone had any problems with such construction? All advice is welcome!

 

[AUCE98]

JAE,

That is a great tip regarding an "envelope" approach to mat design, Thanks.

Auce98

 

[JAE]

Hey - I didn't invent it - just passed down from my own mentors and this is where Eng-Tips can effectively spread lots of good ideas around the engineering community.

 

[J1D]

damo74
I don’t quite understand the 0.01m. Modulus of subgrade is the slope of the stress-displacement curve. If the displacement (or settlement) of the subgrade happens to be 0.01m when the stress reaches to a value 150kPa (it doesn’t matter the stress is allowable or not, once it is significant enough), then the secant modulus is 150/0.01=15000kPa. Am I right?

 

[damo74]

J1D I concur.